In the past year, we investigated the on- or off-pathway nature of protein folding intermediates. It is well known that many proteins fold from their unfolded (U) to the native state (N) through intermediates (I). However, it is not known whether the intermediates are on-pathway (U<-->I<-->N) or off-pathway (I<-->U<-->N) due to misfolding. This is the most elusive problem in the field of protein folding. We have proposed a kinetic criterion that allows us to show an intermediate can be demonstrated to be on-pathway under favorable conditions. Using this criterion, we are able to show that the folding intermediates of hen egg white lysozyme, cytochrome c, and bovine ribonuclese A are on the folding pathways. We also proposed a new model to describe the relationship between hydrogen exchange (HX) behaviors and protein folding pathways. This model illustrates that the commonly used Linderstrom-Lang model is inadequate to describe the HX behavior for a protein with folding intermediate.We solved a long-lasting problem in the folding pathway of cytochrome c. It was widely shown that cytochrome c folds heterogeneously at neutral pH. We demonstrated that such heterogeneous folding behavior is due to intermolecular aggregation.We also solved a highly debated problem on the relationship between native-state HX and protein folding pathways of barnase. We demonstrated that barnase does not fold through an intermediate as commonly perceived. We establisghed a phage-display system which allows us to generate a library of ~1 miilion designed proteins. When combined with proteolysis, we found that the stable protein can be easily selected from the mixture which includes unstable mutants in a control experiment using protein A B-domain. We will use this system to design a helix bundle protein. - protein folding, protein stability, protein structure, - Neither Human Subjects nor Human Tissues